010 GitHub stars
02Sample size and power analysis to ensure experiment validity
03Standardized analysis summaries with clear ship/extend/stop recommendations
04Sample Ratio Mismatch (SRM) detection to flag randomization issues
05Automated statistical significance and p-value calculation via Python scripts
06Guardrail metric monitoring to prevent unintended negative side effects